Robust Online Overdetermined Independent Vector Analysis Based on Bilinear Decomposition

2026-01-18

AI Summary

This work addresses the degradation in estimation accuracy caused by parameter explosion in online blind source separation using large microphone arrays. To mitigate this issue, the authors propose an overdetermined independent vector analysis (OverIVA) method based on bilinear decomposition. By factorizing long separation filters into pairs of shorter filters, the model significantly reduces its parameter count. An alternating iterative projection algorithm is further designed to effectively decouple and update these filter pairs, balancing computational efficiency with numerical stability. The proposed approach achieves superior separation performance and enhanced robustness compared to existing methods, all while substantially reducing the number of model parameters.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Intelligent Robots: State Estimation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Online blind source separation is essential for both speech communication and human-machine interaction. Among existing approaches, overdetermined independent vector analysis (OverIVA) delivers strong performance by exploiting the statistical independence of source signals and the orthogonality between source and noise subspaces. However, when applied to large microphone arrays, the number of parameters grows rapidly, which can degrade online estimation accuracy. To overcome this challenge, we propose decomposing each long separation filter into a bilinear form of two shorter filters, thereby reducing the number of parameters. Because the two filters are closely coupled, we design an alternating iterative projection algorithm to update them in turn. Simulation results show that, with far fewer parameters, the proposed method achieves improved performance and robustness.

Problem

• blind source separation
• overdetermined independent vector analysis
• large microphone arrays
• parameter growth
• online estimation

Innovation

• bilinear decomposition
• overdetermined independent vector analysis
• online blind source separation
• alternating iterative projection
• parameter reduction